Intergroup conflict can be expected to occur whenever animals form groups, and particularly when there is variation among groups in their access to a desired resource. Given that group-living occurs in many species (Krause & Ruxton Reference Krause and Ruxton2002), intergroup conflict is relatively common and is reviewed in Christensen and Radford (Reference Christensen and Radford2018). Indeed, as identified by De Dreu and Gross (D&G) in their thought-provoking article, cooperation and conflict often go hand-in-hand. We applaud D&G for developing new ideas to help further our understanding of intergroup conflict in humans. We wholeheartedly agree with their criticism of existing models of human conflict, which assume symmetric values between defending and attacking groups (sect. 5, para. 1). These assumptions are not realistic, and D&G's focus on understanding asymmetries between groups is an excellent idea.
We, however, feel that the article could have been strengthened by greater consideration of the large body of research on intergroup interactions in nonhuman animals. Indeed, some of the predictions made by D&G have already been tested in animals. For example, research on primates has revealed that coordinated defense is less likely when the attacking individual(s) is targeting a particular individual (for example, the dominant breeding individual; Schindler & Radford Reference Schindler and Radford2018). Subordinate individuals may be less willing to participate in such cases of targeted attack, and opposite-sex individuals do not intervene in fights between same-sex residents and attacking individuals in a number of species (Cant et al. Reference Cant, Otali and Mwanguhya2002; Ridley Reference Ridley, Koenig and Dickinson2016). Therefore, the prediction that individual interests are more aligned in defending groups because they share a common fate when they lose (D&G, sect. 4.1, paras. 2 and 3) is not always supported. When a dominant individual is defeated by an attacker, for example, the opposite-sex group members can still breed with the “replacement” individual, creating a large difference between group members in the “cost” of a successful out-group attack.
The fact that individuals may vary in their contributions to intergroup conflict can also be understood by the collective action problem, where free-riders reap the benefits of successful attack or defense without incurring the costs (Nunn Reference Nunn and Kappeler2000). Recent studies have revealed considerable variation in individual contributions to intergroup conflict (Bonanni et al. Reference Bonanni, Valsecchi and Natoli2010; Mirville et al. Reference Mirville, Ridley, Samedi, Vecellio, Ndagijimana, Stoinski and Grueter2018), and these contributions are strongly related to the individual costs and benefits from investing in conflict. Therefore, the level of variation in benefits between individuals may determine the collective action of the group, and the benefits of investing in conflict should be considered at the individual level. This concept, which recognizes within-group conflict over defense against intruders, has recently been formally modeled by Schindler and Radford (Reference Schindler and Radford2018). Thus, a consideration of the collective action problem would help parameterize the current models of human conflict presented by D&G.
In order to understand intergroup conflict, the motivation to invest in conflict needs to be considered. In nonhuman animals, patterns of conflict are commonly considered in relation to biological parameters. These include the sex and age of the individuals competing, and their access to resources such as food, territory, and reproduction (reviewed in Christensen & Radford Reference Christensen and Radford2018). Often, these biological parameters prove to be good predictors of observed patterns of conflict (Thompson et al. Reference Thompson, Marshall, Vitikainen and Cant2017). Although we appreciate that, at times, it is more difficult to measure some biological parameters in humans, we believe that their inclusion into human conflict research is necessary. D&G state that evidence for the causality of attacks is currently limited (sect. 5.2, para. 4). However, causality is documented in many nonhuman animal species (e.g., Thompson et al. Reference Thompson, Marshall, Vitikainen and Cant2017; Wilson & Wrangham Reference Wilson and Wrangham2003; among others), which can prove very informative for human studies.
Additionally, a considerable body of research on intergroup interactions in primates exists and would have provided an excellent background for some of the ideas presented in D&G's article. Chimpanzees (Pan troglodytes) and humans are considered to have similar social groupings and motivations for conflict (Wrangham & Glowacki Reference Wrangham and Glowacki2012). Thus, the idea by D&G regarding why attack is less successful than defense would have benefitted from a consideration of the “imbalance of power” model, which has been extensively tested in chimpanzee societies (Wilson Reference Wilson2001; Wrangham Reference Wrangham1999). This model assumes that one of the determinants of attack is a sufficient variation in party size between the attacker and the defender (Wrangham Reference Wrangham1999), and research supports this prediction, that intergroup attacks are more likely when the attacking group has numerical superiority (Wilson Reference Wilson2001). Although the idea of asymmetry between social groups was a central topic for D&G, evidence for the imbalance of power model was not addressed.
The prediction that intergroup conflict will affect within-group behavior in humans has been tested in a number of nonhuman animals. These studies provide considerable support for some of the predictions made by D&G. For example, there is considerable evidence that intergroup conflict increases affiliative behaviour between group members in the wood hoopoe (Phoeniculus purpureus) (Radford Reference Radford2008), mountain gorillas (Gorilla beringei beringei) (Mirville Reference Mirville2018), and other primates (Majolo et al. Reference Majolo, de Bortoli Vizoli and Lehmann2016). This suggests that intergroup conflict may have an important influence on the dynamics of group-living behavior and would have been a useful inclusion to suggest that behavioural similarities across the animal kingdom may exist.
In summary, we found the article of D&G an extremely interesting read but felt that the article addressed primarily strategic decisions, without considering the biology behind these decisions. There is a large body of research on social (nonhuman) animal groups that has quantified the costs and benefits of intergroup conflict, and could provide support to many of the ideas put forward in the article. We encourage incorporation of some of this research into human models of intergroup conflict as a productive way to create more realistic theoretical models of asymmetric conflict.
You have
Access
Intergroup conflict can be expected to occur whenever animals form groups, and particularly when there is variation among groups in their access to a desired resource. Given that group-living occurs in many species (Krause & Ruxton Reference Krause and Ruxton2002), intergroup conflict is relatively common and is reviewed in Christensen and Radford (Reference Christensen and Radford2018). Indeed, as identified by De Dreu and Gross (D&G) in their thought-provoking article, cooperation and conflict often go hand-in-hand. We applaud D&G for developing new ideas to help further our understanding of intergroup conflict in humans. We wholeheartedly agree with their criticism of existing models of human conflict, which assume symmetric values between defending and attacking groups (sect. 5, para. 1). These assumptions are not realistic, and D&G's focus on understanding asymmetries between groups is an excellent idea.
We, however, feel that the article could have been strengthened by greater consideration of the large body of research on intergroup interactions in nonhuman animals. Indeed, some of the predictions made by D&G have already been tested in animals. For example, research on primates has revealed that coordinated defense is less likely when the attacking individual(s) is targeting a particular individual (for example, the dominant breeding individual; Schindler & Radford Reference Schindler and Radford2018). Subordinate individuals may be less willing to participate in such cases of targeted attack, and opposite-sex individuals do not intervene in fights between same-sex residents and attacking individuals in a number of species (Cant et al. Reference Cant, Otali and Mwanguhya2002; Ridley Reference Ridley, Koenig and Dickinson2016). Therefore, the prediction that individual interests are more aligned in defending groups because they share a common fate when they lose (D&G, sect. 4.1, paras. 2 and 3) is not always supported. When a dominant individual is defeated by an attacker, for example, the opposite-sex group members can still breed with the “replacement” individual, creating a large difference between group members in the “cost” of a successful out-group attack.
The fact that individuals may vary in their contributions to intergroup conflict can also be understood by the collective action problem, where free-riders reap the benefits of successful attack or defense without incurring the costs (Nunn Reference Nunn and Kappeler2000). Recent studies have revealed considerable variation in individual contributions to intergroup conflict (Bonanni et al. Reference Bonanni, Valsecchi and Natoli2010; Mirville et al. Reference Mirville, Ridley, Samedi, Vecellio, Ndagijimana, Stoinski and Grueter2018), and these contributions are strongly related to the individual costs and benefits from investing in conflict. Therefore, the level of variation in benefits between individuals may determine the collective action of the group, and the benefits of investing in conflict should be considered at the individual level. This concept, which recognizes within-group conflict over defense against intruders, has recently been formally modeled by Schindler and Radford (Reference Schindler and Radford2018). Thus, a consideration of the collective action problem would help parameterize the current models of human conflict presented by D&G.
In order to understand intergroup conflict, the motivation to invest in conflict needs to be considered. In nonhuman animals, patterns of conflict are commonly considered in relation to biological parameters. These include the sex and age of the individuals competing, and their access to resources such as food, territory, and reproduction (reviewed in Christensen & Radford Reference Christensen and Radford2018). Often, these biological parameters prove to be good predictors of observed patterns of conflict (Thompson et al. Reference Thompson, Marshall, Vitikainen and Cant2017). Although we appreciate that, at times, it is more difficult to measure some biological parameters in humans, we believe that their inclusion into human conflict research is necessary. D&G state that evidence for the causality of attacks is currently limited (sect. 5.2, para. 4). However, causality is documented in many nonhuman animal species (e.g., Thompson et al. Reference Thompson, Marshall, Vitikainen and Cant2017; Wilson & Wrangham Reference Wilson and Wrangham2003; among others), which can prove very informative for human studies.
Additionally, a considerable body of research on intergroup interactions in primates exists and would have provided an excellent background for some of the ideas presented in D&G's article. Chimpanzees (Pan troglodytes) and humans are considered to have similar social groupings and motivations for conflict (Wrangham & Glowacki Reference Wrangham and Glowacki2012). Thus, the idea by D&G regarding why attack is less successful than defense would have benefitted from a consideration of the “imbalance of power” model, which has been extensively tested in chimpanzee societies (Wilson Reference Wilson2001; Wrangham Reference Wrangham1999). This model assumes that one of the determinants of attack is a sufficient variation in party size between the attacker and the defender (Wrangham Reference Wrangham1999), and research supports this prediction, that intergroup attacks are more likely when the attacking group has numerical superiority (Wilson Reference Wilson2001). Although the idea of asymmetry between social groups was a central topic for D&G, evidence for the imbalance of power model was not addressed.
The prediction that intergroup conflict will affect within-group behavior in humans has been tested in a number of nonhuman animals. These studies provide considerable support for some of the predictions made by D&G. For example, there is considerable evidence that intergroup conflict increases affiliative behaviour between group members in the wood hoopoe (Phoeniculus purpureus) (Radford Reference Radford2008), mountain gorillas (Gorilla beringei beringei) (Mirville Reference Mirville2018), and other primates (Majolo et al. Reference Majolo, de Bortoli Vizoli and Lehmann2016). This suggests that intergroup conflict may have an important influence on the dynamics of group-living behavior and would have been a useful inclusion to suggest that behavioural similarities across the animal kingdom may exist.
In summary, we found the article of D&G an extremely interesting read but felt that the article addressed primarily strategic decisions, without considering the biology behind these decisions. There is a large body of research on social (nonhuman) animal groups that has quantified the costs and benefits of intergroup conflict, and could provide support to many of the ideas put forward in the article. We encourage incorporation of some of this research into human models of intergroup conflict as a productive way to create more realistic theoretical models of asymmetric conflict.